Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

Commits on main

e55bedd
f239f8c
Simplify default calibration dataset log (#2162)
Wang, Chang committed 3 months ago
960dc7f
03a4429
Add cpu & xpu nightly test (#2149)
Sun, Xuehao committed 3 months ago
436dc24
[ARK] Optimize SYCL-TLA woqgemm_s8 on B70 (#2133)
Zhenzhong Xu committed 3 months ago
2ea60e5
Set the quant_method to auto-round by default (#2138)
Zhenzhong Xu committed 3 months ago
abacdad
Support SVDQuant (#2095)
Wang, Chang committed 3 months ago
ccb3658
refine awq, bugfix (#2115)
Weiwei committed 3 months ago
a3af367
feat: Support world model Cosmos3 (#2040)
Liang Lv committed 3 months ago
9181fa9
cee7109
refine eval, add more option (#2114)
Weiwei committed 3 months ago
fe7676d
Migrate cpu sdpa from NS flash attn (#2092)
jijiaz committed 3 months ago
7402af8
Refactor unit test scripts (#2148)
Sun, Xuehao committed 3 months ago
506bd5f
f17d9cd
fix performance regression bug (#2136)
Heng Guo committed 3 months ago
f8ad63f
Add python 3.14t compatibility tests (#2110)
Sun, Xuehao committed 3 months ago
98f2cdb
support one more mxfp4 variant (#2132)
Wenhua Cheng committed 3 months ago
f706da6
Add llmc test back (#2131)
Yi Liu committed 3 months ago
1f6b215
fix sparge on torch2.13 (#2130)
Yi Liu committed 3 months ago
d8a3005
Refator UT and improve code coverage (#2057)
Liang Lv committed 3 months ago
7a75218
refactor scheme,layer_config and formats (#2013)
Heng Guo committed 3 months ago
77e8ea5